The present invention relates to a method for the preparation and purification of compounds employing a novel support. More particularly, the present invention relates to a method for preparing and purifying a library of compounds useful as pharmaceutical agents using a tetrabenzo[a,c,g,i]fluorene group (Tbf).
A key step in any drug development program is the identification of a lead compound suitable for clinical trials. Increasing competitiveness in the pharmaceutical market and escalating costs for discovery efforts have placed extreme demands on the pharmaceutical industry. The search for new drugs has generally relied on traditional medicinal chemistry approaches which are both time consuming and costly.
The discovery and development of a new drug takes on average 12 years, at an estimated cost of $357 million (Pharmaceutical Manufacturers Association, "Facts at a Glance", 1993). Recently, advances in high throughput synthesis (HTS) and laboratory automation have resulted in the synthesis of medicinal compounds becoming a rate limiting step in the drug discovery process.
Traditional methods for the generation and evaluation of large numbers of compounds have relied heavily on natural products, fermentation broths, marine organisms, and recombinant and chemically synthesized peptides. Combinatorial chemistry is a synthetic strategy by which large, diverse libraries of compounds can be created and has added to the pool of compound sources available at present in the pharmaceutical industry (Fruchtel J. S., Jung G., Angew, Chem. Int. Ed. Engl., 1996;35:17). The advent of combinatorial chemistry has provided a means for the preparation of hundreds, or even thousands of diverse chemical libraries at a fraction of the normal cost and time.
A recent phenomenon, combinatorial chemistry has resulted in the successful preparation of peptides and oligonucleotide based libraries (Gallop M. A., Barrett R. W., Dower W. J., Foder S. P., Gordon E. M., J. Med. Chem., 1994;37:1232). This new field of chemistry has also been expanded to small organic molecules such as the benzodiazepines (Hobbs DeWitt S., Schroeder M. C., Stankovic C. J., Cody D. M. R., Pavia M. R., Proc. Natl. Acad. Sci. USA, 1993;90:6909; Ellman J. A., Bunin B. A., J. Am. Chem. Soc., 1992;114:10997) and hydantoins (Hobbs DeWitt S., Schroeder M. C., Stankovic C. J., Cody D. M. R., Pavia M. R., Proc. Natl. Acad. Sci. USA, 1993;90:6909). Consistent with this trend, most of the top selling drugs on the market are low molecular weight, heterocyclic compounds.
To date, combinatorial strategies have primarily concentrated on solid phase synthesis (SPS) methodologies. Historically, SPS has focused on the preparation on biopolymers such as peptides and oligonucleotides by the use of a few well-characterized chemical transformations to generate repetitive structural backbones. In comparison, SPS of small molecule targets is not fully understood; however, the inherent advantages of the methodology over traditional solution based methods has resulted in the development of solid phase organic synthesis (SPOS).
A variety of solid supports have been described for SPS, most notably the use of functionalised cross-linked polystyrene. Typically, excess reagents are readily tolerated by the solid support, reactions generally show favorable kinetics and can be driven to completion, product isolation is improved by washing away excess reagents from the solid support, and no purification of reaction intermediates is required.
The selection and use of polystyrene resins is, however, dependent upon compatibility with the reaction route necessary to synthesize the target molecule, as well as the method of attachment and cleavage from the solid support. Mechanical and thermal stability of the solid support should also be considered in combination with the method of agitation and the temperature range of the synthetic route. Furthermore, the presence of entrapped impurities and resin by-products can impact the final product yields and purities (MacDonald A. A., DeWitt S. M., Ghosh S., Hogan E. M., Kieras L., Czarnik A. W., Ramage R., Molecular Diversity, 1996;3:183-186).
In order to circumvent the problems associated with solid phase synthesis, there has been some work in the area of generating libraries of compounds using traditional solution phase methodologies. A major drawback of this approach is the general requirement for purification of reaction intermediates which can be both costly and time consuming, particularly if a library of compounds is being prepared.
One method that utilizes the advantages of both solid phase synthesis and solution phase synthesis has been the development of tetrabenzo[a,c,g,i]fluorene (Tbf) and its affinity to porous graphitized carbon (PGC) which was developed by Ramage R., et al., in the late 1980s (Ramage R., Raphy G., Tetrahedron Lett., 1992;33:385).
The general concept involves tagging the N-terminus of a peptide chain with a suitable aromatic derivatized protecting group. Subsequently, the tag is selectively adsorbed on to PGC. Truncated peptides are then removed by washing, and the final pure peptide can be obtained by deprotection and elution from PGC.
The introduction of solid phase peptide synthesis (SPPS) by Merrifield in 1963 greatly simplified the task of peptide synthesis. Yet one of the main obstacles of SPPS is the difficulty of purification of the final product due to the accumulation of truncated peptides on the resin. Several chromatographic separation methods are available for purification, one of which involves affinity based purification with PGC.
PGC has a porous two-dimensional graphite structure with a large surface area available for affinity binding. More importantly, PGC has strong hydrophobic adsorption with unique selectivity, particularly to aromatic systems. In order to exploit this hydrophobic interaction which PGC exhibited for large, flat molecules, Ramage and Raphy, set out to design a new, planar, aromatic system for the purification of peptides (Ramage R., Raphy C., Tetrahedron Lett., 1992;33:385).
In 1960, Martin, et al., reported the first synthesis of Tbf (de Ridder R., Martin R. H., Bull. Soc. Chim Belg., 1960;69:534); however, the chemistry of Tbf received little attention until 1988 when Ramage and Raphy synthesized three derivatives of Tbf as potential intermediates in the synthesis of N-Tbfmoc amino acids and peptides. In addition to the strong fluorescent properties of Tbf, it also exhibited strong hydrophobic binding to PGC. Ramage, et al., have successfully applied this technique to the design of a highly hydrophobic N-amino protecting group and a 5'-hydroxyl protecting group for peptide and DNA synthesis, respectively. It was found to be useful in the final purification step either by selective binding to PGC, or as a hydrophobic chromatographic label to allow HPLC based purification.
A key component in the synthesis of appropriate protecting groups has been the reactivity of the methylene bridge at position 17 of the Tbf molecule. The hydrogen atoms are highly acidic which is attributed to the high stability of the resonance stabilized cyclopentadienyl anion generated under basic conditions.
With these unique properties in mind, we have surprisingly and unexpectedly found that by utilizing the charcoal/Tbf affinity interaction such an approach could be used in the generation of compounds and in particular a library of compounds using, for example, the Diversomer.RTM. Technology disclosed in U.S. Pat. No. 5,324,483 which is herein incorporated by reference. Analogous to SPOS, such a system can be used to replace the standard polystyrene based solid supports with charcoal. Thus, the chemistry of Tbf derivatives are modified depending on the synthetic scheme, and all reactions are carried out in solution phase. While this eliminates the problems associated with SPS, the use of charcoal as a solid support could be introduced at the end of each stage in the synthesis to purify the intermediates and final product(s).